Nirmatrelvir-Ritonavir Shows No Benefit for Long COVID

Nirmatrelvir-ritonavir (Paxlovid) provided no measurable benefit for patients with long COVID, according to results from RECOVER-VITAL (NCT05595369), a phase 2 randomized, double-blind, placebo-controlled trial published in The Lancet Infectious Diseases.¹

The trial enrolled 959 participants across 69 US sites between July 2023 and September 2024 to test whether 15 or 25 days of the antiviral could improve symptoms in 3 of the most common long COVID phenotypes: cognitive dysfunction, autonomic dysfunction, and exercise intolerance. Neither treatment duration outperformed placebo on any primary end point, and no new safety signals emerged.¹

Trial Design Targeted Viral Persistence

Long COVID, also known as post-acute sequelae of SARS-CoV-2 infection, affects more than 5% of US adults, with more than 1% experiencing significant activity limitations, according to the study authors. Among the leading hypotheses for its cause is viral persistence, the idea that ongoing low-level or compartmentalized SARS-CoV-2 replication drives continued symptoms, an idea supported by reports of virus detection years after primary infection in some patients with long COVID.¹

RECOVER-VITAL was designed within the National Institutes of Health RECOVER (Researching COVID to Enhance Recovery) initiative to test that hypothesis directly using nirmatrelvir-ritonavir, an antiviral combination already used against acute SARS-CoV-2 infection.¹

Of 1207 individuals screened, 964 were randomly allocated and 959 initiated treatment, enrolled as 332 participants in the cognitive phenotype, 334 in the autonomic phenotype, and 332 in the exercise phenotype. Participants were 67% female, 78% white, and 11% Hispanic, Latino, or Spanish, with a median age of 49 years.¹

Within each phenotype, participants were randomized 1:1:1 to 15 days of nirmatrelvir-ritonavir (300 mg nirmatrelvir/100 mg ritonavir twice daily) followed by 10 days of placebo, 25 days of the same active regimen, or 25 days of low-dose ritonavir alone, which was used in the control arm specifically to preserve blinding because of the drug's distinctive taste. The primary end point was a clinically meaningful improvement at day 90 on a phenotype-specific patient-reported outcome measure, and secondary end points relied on objective performance measures such as cognitive testing and a walking-endurance test.¹

No Improvement Across Any Phenotype

None of the primary comparisons reached statistical significance. In the cognitive phenotype, 58% of participants receiving the 25-day regimen and 53% receiving the 15-day regimen showed meaningful improvement, compared with 55% receiving placebo, for adjusted differences of 3.2% (95% CI, -10.4 to 16.8; P = .65) and -2.2% (95% CI, -15.5 to 11.1; P = .74), respectively.¹

In the autonomic phenotype, improvement occurred in 62% of the 25-day group, 70% of the 15-day group, and 70% of the placebo group, for adjusted differences of -6.4% (95% CI, -18.5 to 5.7; P = .30) and -0.1% (95% CI, -12.5 to 12.3; P = .99). In the exercise phenotype, 25% of the 25-day group, 34% of the 15-day group, and 33% of the placebo group met the improvement threshold, for adjusted differences of -7.8% (95% CI, -19.5 to 3.8; P = .19) and 0.9% (95% CI, -11.4 to 13.2; P = .88).¹

Secondary end points based on objective performance measures told a similar story, with one notable exception. On the exercise phenotype's performance-based measure of post-exertional malaise, the 25-day regimen showed an adjusted difference of -13.8% (95% CI, -24.9 to 2.8; P = .014) relative to placebo, a statistically significant result favoring the placebo group rather than the active drug. The study authors did not interpret this as evidence of harm from treatment, noting that comparisons across all end points and time points, including day 180, showed consistent findings of no benefit rather than a clear signal in either direction.¹

"Nirmatrelvir-ritonavir for 15 days or 25 days showed no evidence of benefit in long COVID in any of the three phenotypes studied," the study authors wrote.¹ "These findings suggest additional approaches to measuring the symptom burden and treating Long COVID are needed."

Safety Findings Were Consistent

There were no deaths during the trial, and 52 serious adverse events occurred among 42 of 963 participants (4%) who received at least 1 dose of study drug, a rate similar across all three treatment groups. Any adverse event was reported by 66% of the 25-day group, 64% of the 15-day group, and 58% of the placebo group; diarrhea and nausea were the most common, occurring more frequently in the active-treatment groups than in the placebo group.¹

The study drug was discontinued because of an adverse event in 3% of the 25-day group, 6% of the 15-day group, and 4% of the placebo group. Investigators considered 3 serious adverse events as treatment related: suspected drug-induced liver injury in a participant on the 25-day regimen, elevated liver function tests in a participant on the 15-day regimen, and drug intolerance in a participant on placebo-ritonavir.¹

The trial's authors noted 2 earlier, smaller randomized trials of 15-day nirmatrelvir-ritonavir, known as STOP-PASC and PAX-LC, also failed to show a benefit but were not considered definitive because of limited enrollment (a combined 151 participants randomized to active drug), broad entry criteria, and outcome measures that were not specific to individual symptom phenotypes. RECOVER-VITAL was designed to address those limitations with a larger sample, phenotype-specific assessment tools, and both patient-reported and performance-based outcomes.¹

A Mixed and Evolving Evidence Base

Nirmatrelvir-ritonavir, marketed as Paxlovid, is used to treat mild to moderate COVID-19 in nonhospitalized patients at high risk of progressing to severe disease, and treatment must begin within 5 days of symptom onset. The standard regimen is 300 mg nirmatrelvir plus 100 mg ritonavir taken twice daily for 5 days, and the combination carries an extensive list of drug interactions and is not recommended for patients with severe kidney or liver disease, factors that pharmacists should continue to screen for regardless of the indication being treated.²

Taste alteration is among the most commonly reported adverse effects, the same property investigators relied on to help mask treatment assignment in RECOVER-VITAL's low-dose ritonavir control arm.^1,2

The RECOVER-VITAL results arrive alongside additional data complicating the drug's evidence base for acute COVID-19 as well. In the PANORAMIC and CanTreatCOVID trials, published in the New England Journal of Medicine, nirmatrelvir-ritonavir did not reduce the combined risk of hospitalization or death among vaccinated, higher-risk outpatients with acute SARS-CoV-2 infection.³

Among 3516 participants in the United Kingdom's PANORAMIC trial, 14 of 1698 (0.8%) participants receiving nirmatrelvir-ritonavir were hospitalized or died, compared with 11 of 1673 (0.7%) receiving usual care. In Canada's CanTreatCOVID trial, 2 of 343 (0.6%) participants receiving nirmatrelvir-ritonavir were hospitalized or died, compared with 4 of 324 (1.2%) receiving usual care. Those results contrast with the earlier EPIC-HR trial, which found a benefit in unvaccinated high-risk outpatients and led to nirmatrelvir-ritonavir's recommendation as first-line therapy for that population.³

Not all recent findings point the same direction, however. A retrospective study published in the Journal of Human Immunity examined 2740 patients with primary or secondary immunodeficiency who had a documented SARS-CoV-2 infection at Lahey Hospital and Medical Center between 2015 and 2024.⁴

After adjusting for confounders including vaccination status, immunodeficiency severity, and acute COVID-19 severity, nirmatrelvir-ritonavir use was associated with a reduced risk of long COVID in this population (OR, 0.43; 95% CI, 0.24-0.72; P = .002), a finding that held across subgroup and sensitivity analyses. The study's authors noted that immunodeficient patients, who face elevated baseline risk for both severe COVID-19 and long COVID, have largely been excluded from randomized long COVID trials, limiting how directly the two sets of findings can be compared.⁴

What It Means for Pharmacists

For pharmacists, RECOVER-VITAL reinforces that no antiviral therapy has yet demonstrated clinical benefit for long COVID in a well-powered, phenotype-specific randomized trial, and patients asking about nirmatrelvir-ritonavir for that purpose should be counseled accordingly. Because the drug is already widely dispensed for acute COVID-19, pharmacists remain the front line for screening the extensive drug interactions and renal or hepatic contraindications that apply whether the antiviral is used for its approved indication or considered off-label for long COVID.^1,2

The trial's investigators also collected blood specimens to test for baseline and on-treatment viral antigen levels; those virologic analyses are still pending and could eventually clarify whether a subgroup of patients with confirmed viral persistence responds differently to antiviral therapy.¹

REFERENCES

1. Baden LR, Shah NS, Liu STH, et al. Nirmatrelvir-ritonavir targeting viral persistence in post-COVID-19 condition (long COVID) in the USA (RECOVER-VITAL): a randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Infect Dis. Published online August 31, 2026. doi:10.1016/S1473-3099(26)00406-8

2. Mayo Clinic. Nirmatrelvir and ritonavir (oral route) description. Mayo Clinic. Accessed September 8, 2026. https://www.mayoclinic.org/drugs-supplements/nirmatrelvir-and-ritonavir-oral-route/description/drg-20528231

3. Butler CC, Pinto AD, Harris V, et al. Oral Nirmatrelvir-Ritonavir for Covid-19 in Higher-Risk Outpatients. N Engl J Med. 2026;394(16):1583-1594. doi:10.1056/NEJMoa2502457

4. Gilbert KM, Aglan M, Jogdand A, et al. Nirmatrelvir/ritonavir use reduces risk for long COVID in patients with immunodeficiency. J Hum Immun. 2025;2(1):e20250107. Published 2025 Nov 4. doi:10.70962/jhi.20250107